Personality feature recognition method and device, equipment, storage medium and product
By extracting the image and voice data when driving a car, combining driving data, and using machine learning algorithms to train personality feature recognition models, the problem of insufficient accuracy of personality feature recognition in the existing technology is solved, and accurate analysis and implementation of driving Personalized application.
Patent Information
- Application Number
- CN202410983811.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art is difficult to accurately identify the personality characteristics of users, which affects the implementation of personalized marketing and intelligent interaction functions.
By obtaining image and voice data when driving a car, extracting action and sound features, combining driving data, generating sample data, using machine learning algorithms to train personality feature recognition models to identify users' personality characteristics.
It improves the accuracy of personality trait recognition, expands the application scenario, and can analyze user personality traits based on driving behavior during the user's driving process.
Smart Images

Figure CN118940130A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of machine learning technology, and in particular to a method, device, equipment, storage medium and product for identifying personality traits. Background Art
[0002] At present, whether in the automotive industry or in other industries, understanding the personality characteristics of users plays an important role in product marketing, intelligent push notifications during sales, and personalized after-sales services. First, understanding the personality characteristics of users can optimize marketing strategies and achieve precise marketing strategies that vary from person to person. Second, understanding the personality characteristics of users can define and upgrade product functions according to user needs. Third, understanding the personality characteristics of users can provide personalized after-sales services based on the user's personality and improve brand maintenance. Fourth, it is also possible to improve personalized intelligent interaction functions and enhance user experience by identifying the personality characteristics of users. Summary of the invention
[0003] The embodiments of the present application provide a method, device, equipment, storage medium and product for identifying personality traits, which can improve the accuracy of personality trait identification. The technical solution is as follows:
[0004] In one aspect, a method for identifying personality traits is provided, the method comprising:
[0005] Based on the questionnaire information, determine the personality characteristics of the sample users;
[0006] Acquire multiple frames of first image data and multiple frames of first voice data of the sample user in the process of driving a car;
[0007] Based on the multiple frames of first image data, determining multiple groups of action features of the sample user in the process of driving a car, each group of action features including head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, upper and lower opening and closing distance of the mouth, and left and right opening and closing distance of the mouth;
[0008] Based on the multiple groups of action features, determine the maximum value, minimum value, variance, mean and median of each action feature included in the multiple groups of action features to obtain the action features of the sample user in the process of driving a car;
[0009] Based on the multiple frames of first voice data, determining multiple groups of voice features of the sample user in the process of driving a car, each group of voice features including pitch, sound intensity, formant, effective voice length, speech speed and number of pauses;
[0010] Based on the multiple groups of sound features, determine the maximum value, minimum value, variance, mean and median of each sound feature included in the multiple groups of sound features to obtain the sound features of the sample user in the process of driving the car;
[0011] Acquire first driving data of the sample user driving a car within a recent preset time period and first vehicle data of the car, and second driving data of the sample user driving a car within a recent preset mileage and second vehicle data of the car;
[0012] Determining variable features corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data;
[0013] Generate sample data based on the personality feature, the action feature, the sound feature and the variable feature, wherein the label of the sample data is the personality feature, and the features of the sample data include the action feature, the sound feature and the variable feature;
[0014] Based on the sample data, a personality feature recognition model is trained by a machine learning algorithm, and the personality feature recognition model is used to recognize the personality features of the user.
[0015] In a possible implementation manner, determining the variable features corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data includes:
[0016] Determine a first set of driving characteristics of the vehicle based on the first driving data, the first set of driving characteristics comprising charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average lateral speed, and maximum steering wheel speed;
[0017] Determine a first set of vehicle characteristics of the vehicle based on the first vehicle data, the first set of vehicle characteristics comprising an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the vehicle;
[0018] Determining a second set of driving characteristics of the vehicle based on the second driving data, the second set of driving characteristics comprising charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average lateral speed, and maximum steering wheel speed;
[0019] determining a second set of vehicle characteristics of the vehicle based on the second vehicle data, the second set of vehicle characteristics comprising an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the vehicle;
[0020] The variable characteristics are determined based on the first set of driving characteristics, the first set of vehicle characteristics, the second set of driving characteristics, and the second set of vehicle characteristics.
[0021] In another possible implementation, determining the variable feature based on the first group of driving features, the first group of vehicle features, the second group of driving features, and the second group of vehicle features includes:
[0022] combining the first set of driving characteristics and the second set of driving characteristics into a driving characteristic;
[0023] When the driving characteristics include two identical characteristics, deleting one of the two characteristics;
[0024] combining the first set of vehicle features and the second set of vehicle features into a vehicle feature;
[0025] When the vehicle features include two identical features, deleting one of the two features;
[0026] The driving characteristics and the vehicle characteristics are combined into the variable characteristics.
[0027] In a possible implementation, determining, based on the multiple frames of first image data, multiple groups of action features of the sample user in a process of driving a car includes:
[0028] For each frame of first image data, performing face detection on the first image data to obtain a face image of the sample user;
[0029] Based on the facial image, determining facial feature points of the sample user;
[0030] Based on the facial feature points, a set of action features of the sample user during driving a car is determined.
[0031] In another possible implementation, the motion feature includes a head swing amplitude;
[0032] The step of determining a set of action features of the sample user in the process of driving a car based on the facial feature points includes:
[0033] Determine first coordinate information of the facial feature point in the world coordinate system and second coordinate information of the facial feature point in the camera coordinate system;
[0034] Based on the first coordinate information and the second coordinate information, determining the head swing amplitude of the sample user during driving the car by using the following formula 1;
[0035] Formula 1:
[0036] in, Represents the second coordinate information, It represents the first coordinate information, and [RT] represents the amplitude of head swing.
[0037] In another possible implementation, the motion features include a left eye opening and closing distance, a right eye opening and closing distance, a mouth upper and lower opening and closing distance, and a mouth left and right opening and closing distance;
[0038] The step of determining a set of action features of the sample user in the process of driving a car based on the facial feature points includes:
[0039] Determine a plurality of left eye key points, a plurality of right eye key points, a plurality of upper and lower mouth feature points, and a plurality of left and right mouth feature points from the facial feature points;
[0040] Determine third coordinate information of the plurality of left eye key points in the camera coordinate system, fourth coordinate information of the plurality of right eye key points in the camera coordinate system, fifth coordinate information of the plurality of upper and lower mouth feature points in the camera coordinate system, and sixth coordinate information of the plurality of left and right mouth feature points in the camera coordinate system;
[0041] Determining the opening and closing distance of the left eye of the sample user during driving the car based on the third coordinate information of the multiple left eye key points;
[0042] Determining the right eye opening and closing distance of the sample user during driving the car based on the fourth coordinate information of the plurality of right eye key points;
[0043] Determine the upper and lower opening and closing distance of the mouth of the sample user when driving the car based on the fifth coordinate information of the plurality of upper and lower mouth feature points;
[0044] Based on the sixth coordinate information of the multiple left and right mouth feature points, the left and right opening and closing distance of the mouth of the sample user when driving a car is determined.
[0045] In another possible implementation, the sound feature includes pitch;
[0046] The determining, based on the multiple frames of first voice data, multiple groups of sound features of the sample user in the process of driving a car includes:
[0047] For each frame of first speech data, determining a fundamental frequency value of the first speech data;
[0048] Based on the fundamental frequency value of the first voice data, the pitch of the sample user determined based on the first voice data is determined by the following formula 2:
[0049] Formula 2:
[0050] Among them, h represents the pitch of the sample user determined based on the first voice data, and f represents the fundamental frequency value of the first voice data.
[0051] In another possible implementation, the sound feature includes sound intensity;
[0052] The determining, based on the multiple frames of first voice data, multiple groups of sound features of the sample user in the process of driving a car includes:
[0053] For each frame of first voice data, performing Fourier transform on the first voice data to obtain a frequency domain signal of the first voice data, wherein the frequency domain signal includes multiple frequencies;
[0054] For each frequency in the frequency domain signal, determine the square of the amplitude of the frequency to obtain an energy spectrum of the frequency;
[0055] Determining a sound pressure value of the first speech data based on the energy spectrum of each frequency;
[0056] Based on the sound pressure value of the first voice data, the voice intensity of the sample user determined based on the first voice data is determined by the following formula 3:
[0057] Formula 3:
[0058] Wherein, Lp represents the voice intensity of the sample user determined based on the first voice data, p represents the sound pressure value, and p0 represents the reference sound pressure value.
[0059] In another possible implementation, the sound feature includes a formant;
[0060] The determining, based on the multiple frames of first voice data, multiple groups of sound features of the sample user in the process of driving a car includes:
[0061] Performing pre-emphasis processing on the multiple frames of first voice data;
[0062] Performing windowing and framing processing on the multiple frames of first voice data after pre-emphasis processing to obtain multiple frames of second voice data;
[0063] The frequency domain signals of the multiple frames of second voice data are obtained by performing Fourier transform on the multiple frames of second voice data through the following formula 4:
[0064] Formula 4:
[0065] Among them, X i (n) represents the frequency domain signal of the second speech data of the i-th frame, x i(n) represents the second voice data of the i-th frame, N represents the frame length of the frame processing, and n represents the number of frames of the second voice data;
[0066] Performing an inverse Fourier transform on the frequency domain signals of the multiple frames of second voice data to obtain an inverse Fourier transform sequence of the multiple frames of second voice data;
[0067] Based on the dop sequence of the multiple frames of second voice data, the envelope of the frequency domain signal of the multiple frames of second voice data is determined by the following formula 5:
[0068] Formula 5:
[0069] Among them, H t (k) represents X i (n) envelope, h i (n) represents the product of the window function and the cepstrum sequence, N represents the frame length of the frame processing, and n represents the number of frames of the second speech data;
[0070] For each second voice data, a maximum value is determined on the envelope of the frequency domain signal of the second voice data to obtain the resonance peak of the sample user determined based on the second voice data.
[0071] In another aspect, a personality feature recognition device is provided, the device comprising:
[0072] A first determination module is used to determine the personality characteristics of sample users based on the questionnaire survey information;
[0073] A first acquisition module, used to acquire multiple frames of first image data and multiple frames of first voice data of the sample user in the process of driving a car;
[0074] A second determination module is used to determine multiple groups of action features of the sample user in the process of driving a car based on the multiple frames of first image data, each group of action features including head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, upper and lower opening and closing distance of the mouth, and left and right opening and closing distance of the mouth;
[0075] A third determination module is used to determine the maximum value, minimum value, variance, mean and median of each action feature included in the multiple groups of action features based on the multiple groups of action features, so as to obtain the action features of the sample user in the process of driving a car;
[0076] a fourth determination module, configured to determine, based on the multiple frames of first voice data, multiple groups of voice features of the sample user in the process of driving a car, each group of voice features including pitch, sound intensity, formant, effective voice length, speech speed, and number of pauses;
[0077] a fifth determination module, configured to determine, based on the multiple groups of sound features, a maximum value, a minimum value, a variance, a mean value, and a median value of each sound feature included in the multiple groups of sound features, and obtain the sound features of the sample user in the process of driving a car;
[0078] A second acquisition module is used to acquire first driving data of the sample user driving a car within a recent preset time period and first vehicle data of the car, as well as second driving data of the sample user within a recent preset mileage and second vehicle data of the car;
[0079] a sixth determination module, configured to determine a variable feature corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data;
[0080] A generating module, configured to generate sample data based on the personality feature, the action feature, the sound feature and the variable feature, wherein the label of the sample data is the personality feature, and the features of the sample data include the action feature, the sound feature and the variable feature;
[0081] The training module is used to train a personality feature recognition model based on the sample data through a machine learning algorithm, and the personality feature recognition model is used to recognize the personality features of the user.
[0082] In a possible implementation, the sixth determination module is used to determine a first group of driving characteristics of the car based on the first driving data, the first group of driving characteristics including charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average value of acceleration, average value of lateral speed, and maximum steering wheel speed; determine a first group of vehicle characteristics of the car based on the first vehicle data, the first group of vehicle characteristics including an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the car; determine a second group of driving characteristics of the car based on the second driving data, the second group of driving characteristics including charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average value of acceleration, average value of lateral speed, and maximum steering wheel speed; determine a second group of vehicle characteristics of the car based on the second vehicle data, the second group of vehicle characteristics including an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the car; determine the variable characteristics based on the first group of driving characteristics, the first group of vehicle characteristics, the second group of driving characteristics, and the second group of vehicle characteristics.
[0083] In another possible implementation, the sixth determination module is used to combine the first group of driving characteristics and the second group of driving characteristics into driving characteristics; when the driving characteristics include two identical characteristics, delete one of the two characteristics; combine the first group of vehicle characteristics and the second group of vehicle characteristics into vehicle characteristics; when the vehicle characteristics include two identical characteristics, delete one of the two characteristics; combine the driving characteristics and the vehicle characteristics into the variable characteristics.
[0084] In another possible implementation, the second determination module is used to perform face detection on each frame of first image data to obtain a face image of the sample user; determine facial feature points of the sample user based on the facial image; and determine a set of action features of the sample user in the process of driving a car based on the facial feature points.
[0085] In another possible implementation, the action feature includes a head swing amplitude; the second determination module is used to determine first coordinate information of the facial feature point in a world coordinate system and second coordinate information of the facial feature point in a camera coordinate system; based on the first coordinate information and the second coordinate information, the head swing amplitude of the sample user in the process of driving a car is determined by the following formula 1;
[0086] Formula 1:
[0087] in, Represents the second coordinate information, It represents the first coordinate information, and [RT] represents the amplitude of head swing.
[0088] In another possible implementation, the motion features include a left eye opening and closing distance, a right eye opening and closing distance, a mouth upper and lower opening and closing distance, and a mouth left and right opening and closing distance;
[0089] The second determination module is used to determine multiple left eye key points, multiple right eye key points, multiple upper and lower mouth feature points and multiple left and right mouth feature points from the facial feature points; determine the third coordinate information of the multiple left eye key points in the camera coordinate system, the fourth coordinate information of the multiple right eye key points in the camera coordinate system, the fifth coordinate information of the multiple upper and lower mouth feature points in the camera coordinate system and the sixth coordinate information of the multiple left and right mouth feature points in the camera coordinate system; based on the third coordinate information of the multiple left eye key points, determine the opening and closing distance of the left eye of the sample user in the process of driving the car; based on the fourth coordinate information of the multiple right eye key points, determine the opening and closing distance of the right eye of the sample user in the process of driving the car; based on the fifth coordinate information of the multiple upper and lower mouth feature points, determine the upper and lower opening and closing distance of the mouth of the sample user in the process of driving the car; based on the sixth coordinate information of the multiple left and right mouth feature points, determine the left and right opening and closing distance of the mouth of the sample user in the process of driving the car.
[0090] In another possible implementation, the sound feature includes pitch;
[0091] The fourth determination module is configured to determine the fundamental frequency value of the first voice data for each frame of the first voice data; and determine the pitch of the sample user determined based on the first voice data by the following formula 2 based on the fundamental frequency value of the first voice data:
[0092] Formula 2:
[0093] Among them, h represents the pitch of the sample user determined based on the first voice data, and f represents the fundamental frequency value of the first voice data.
[0094] In another possible implementation, the sound feature includes sound intensity;
[0095] The fourth determination module is used to perform Fourier transform on each frame of first voice data to obtain a frequency domain signal of the first voice data, wherein the frequency domain signal includes multiple frequencies; for each frequency in the frequency domain signal, determine the square of the amplitude of the frequency to obtain an energy spectrum of the frequency; determine the sound pressure value of the first voice data based on the energy spectrum of each frequency; and determine the sound intensity of the sample user determined based on the first voice data based on the sound pressure value of the first voice data by the following formula three:
[0096] Formula 3:
[0097] Wherein, Lp represents the voice intensity of the sample user determined based on the first voice data, p represents the sound pressure value, and p0 represents the reference sound pressure value.
[0098] In another possible implementation, the sound feature includes a formant;
[0099] The fourth determination module is used to perform pre-emphasis processing on the multiple frames of first voice data; perform windowing and framing processing on the multiple frames of first voice data after pre-emphasis processing to obtain multiple frames of second voice data; and perform Fourier transform on the multiple frames of second voice data according to the following formula 4 to obtain frequency domain signals of the multiple frames of second voice data:
[0100] Formula 4:
[0101] Among them, X i (n) represents the frequency domain signal of the second speech data of the i-th frame, x i (n) represents the second voice data of the i-th frame, N represents the frame length of the frame processing, and n represents the number of frames of the second voice data;
[0102] Performing an inverse Fourier transform on the frequency domain signals of the multiple frames of second voice data to obtain an inverted Fourier transform sequence of the multiple frames of second voice data; and determining the envelope of the frequency domain signals of the multiple frames of second voice data by the following formula 5 based on the inverted Fourier transform sequence of the multiple frames of second voice data:
[0103] Formula 5:
[0104] Among them, H i (k) represents X i (n) envelope, h i (n) represents the product of the window function and the cepstrum sequence, N represents the frame length of the frame processing, and n represents the number of frames of the second speech data;
[0105] For each second voice data, a maximum value is determined on the envelope of the frequency domain signal of the second voice data to obtain the resonance peak of the sample user determined based on the second voice data.
[0106] On the other hand, a computer device is provided, comprising a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the above-mentioned personality feature recognition method.
[0107] On the other hand, a computer-readable storage medium is provided, wherein at least one program code is stored in the storage medium, and the at least one program code is loaded and executed by a processor to implement the above-mentioned personality feature recognition method.
[0108] On the other hand, a computer program product is provided, wherein the product stores at least one program code, and the at least one program code is used to be executed by a processor to implement the above-mentioned personality feature recognition method.
[0109] In the embodiment of the present application, since the action characteristics, sound characteristics and variable characteristics of the user in the process of driving a car are associated with the personality characteristics to generate sample data, the personality characteristic recognition model trained based on the sample data can analyze the user's personality characteristics based on the user's driving behavior during the user's driving process, which not only expands the application scenarios of personality characteristic recognition, but also improves the accuracy of personality characteristic recognition.
[0110] It is to be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0111] Figure 1 is a schematic diagram of a development process of a method for identifying personality traits according to an exemplary embodiment of the present application;
[0112] Figure 2 is a flow chart of a method for identifying personality traits shown in an exemplary embodiment of the present application;
[0113] Figure 3 is a flow chart of a method for identifying personality traits shown in an exemplary embodiment of the present application;
[0114] Figure 4 is a block diagram of a personality feature recognition device shown in an exemplary embodiment of the present application;
[0115] Figure 5 It is a block diagram of a computer device shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0116] In order to make the technical solutions and advantages of the present application clearer, the implementation methods of the present application are described in further detail below.
[0117] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0118] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions. For example, the questionnaire information, first image data, first voice data, first driving data, first vehicle data, second driving data and second vehicle data involved in this application are all obtained with full authorization.
[0119] Please refer to Figure 1 , which shows a schematic diagram of the development process of a personality feature recognition method shown in an exemplary embodiment of the present application; the development process includes determining the personality features of the sample user based on the questionnaire information. The microphone obtains multiple frames of first voice data of the sample user in the process of driving the car, and transmits the multiple frames of first voice data to the local hard disk via USB (Universal Serial Bus). The local hard disk determines the sound features of the sample user through Algorithm 1 based on the multiple frames of first voice data. The camera obtains multiple frames of first image data of the sample user in the process of driving the car, and transmits the multiple frames of first image data to the local hard disk via USB. The local hard disk determines the action features of the sample user through Algorithm 2 based on the multiple frames of first image data. The in-vehicle terminal obtains the first driving data of the sample user driving the car within a recent preset time and the first vehicle data of the car, as well as the second driving data and the second vehicle data of the vehicle within a recent preset mileage, and transmits the first driving data, the first vehicle data, the second driving data and the second vehicle data to the big data platform by wireless transmission. The big data platform determines the variable characteristics through Algorithm 3 based on the first driving data, the first vehicle data, the second driving data and the second vehicle data. The computer device is embedded with a personality feature recognition algorithm. The computer device performs model training through the personality feature recognition algorithm based on personality features, action features, sound features and variable features to obtain a personality feature recognition model. The personality feature recognition model is used to identify the personality features of the user.
[0120] Please refer to Figure 2 , which shows a flowchart of a method for identifying personality characteristics according to an exemplary embodiment of the present application. Figure 2 , the method comprising:
[0121] Step 201: Determine the personality characteristics of sample users based on the questionnaire survey information.
[0122] Prior to this step, a first personality test form is issued to the sample user. The first personality test form includes multiple questions and multiple options corresponding to each question. The sample user can determine the target option that meets his or her personality characteristics from the multiple options corresponding to each question, and then mark the target option corresponding to each question in the personality test form. For the sake of distinction, the personality test form with the target option corresponding to each question marked by the sample user is called the second personality test form. In this step, the computer device obtains each question and the answer information corresponding to each question in the second personality test form, obtains the questionnaire survey information, and then determines the personality characteristics of the sample user based on the questionnaire survey information.
[0123] The number of sample users is multiple, and the number of sample users is greater than the preset number, that is, the number of sample users is guaranteed to be sufficient, so as to improve the accuracy of training the personality feature recognition model based on the personality features of multiple sample users. In addition, before issuing the first personality test form to the sample user, a privacy disclosure agreement is signed with the sample user, and a subsidy is issued to the sample user, so that the sample user can take the questionnaire seriously, thereby improving the quality of the obtained questionnaire information.
[0124] Step 202: Acquire multiple frames of first image data and multiple frames of first voice data of a sample user while driving a car.
[0125] The car includes a camera and a microphone. The computer device obtains multiple frames of first image data of a sample user during driving the car through the camera, and obtains multiple frames of first voice data of the sample user during driving the car through the microphone.
[0126] After the computer device acquires multiple frames of first image data, it needs to determine the facial image of the sample user based on the multiple frames of first image data; therefore, after the computer device acquires multiple frames of first image data, for each frame of first image data, it determines whether the first image data includes the facial image of the sample user; if the facial image of the sample user is included in the first image data, step 203 is executed; if the facial image of the sample user is not included in the first image data, the first image data is re-acquired until the first image data including the facial image is acquired. In addition, the computer device may also determine whether the image quality of the first image data meets the quality condition. If the image quality of the first image data meets the quality condition, step 203 is executed; if the image quality of the first image data does not meet the quality condition, the first image data is re-acquired until the first image data meeting the quality condition is acquired.
[0127] Since the embodiment of the present application is to obtain the correlation between the sound characteristics of the sample user when driving and the Big Five personality, and since the car may include other users in the co-pilot seat or the back seat in addition to the sample user in the main driving seat; therefore, it is necessary to determine whether the first voice data obtained is the voice data of the sample user. If the first voice data is the voice data of the sample user, execute step 205; if the first voice data is not the voice data of the sample user, re-acquire the first voice data until the voice data of the sample user is obtained; in addition, if the first voice data includes the voice data of the sample user, but also includes the voice data of other users, it is still necessary to re-acquire the first voice data, so as to obtain the first voice data that only includes the voice data of the sample user, thereby improving the quality of the obtained first voice data.
[0128] The voiceprint features of the sample user are stored in advance in the computer device; accordingly, the step for the computer device to determine whether the acquired first voice data is the voice data of the sample user may be: the computer device extracts the voiceprint features of the first voice data, and determines whether the extracted voiceprint features match the voiceprint features of the sample user; when the extracted voiceprint features match the voiceprint features of the sample user, it is determined that the first voice data is the voice data of the sample user; when the extracted warming features do not match the voiceprint features of the sample user, it is determined that the first voice data is not the voice data of the sample user.
[0129] Step 203: Based on the multiple frames of first image data, multiple groups of motion features of the sample user in the process of driving a car are determined, each group of motion features includes the head swing amplitude, the left eye opening and closing distance, the right eye opening and closing distance, the upper and lower opening and closing distance of the mouth, and the left and right opening and closing distance of the mouth.
[0130] Because we need to find the correlation between the action characteristics of the sample users when driving and the Big Five personality traits, we need to develop an algorithm to solve the head swing amplitude, the opening and closing distance of the left and right eyes, and the opening and closing distance of the mouth when driving, and then solve the maximum, minimum, variance, mean and median of these characteristics to obtain the action characteristics of the sample users.
[0131] One frame of first image data corresponds to a set of action features; this step can be implemented by the following steps (1) to (3), including:
[0132] (1) For each frame of first image data, face detection is performed on the first image data to obtain a face image of a sample user.
[0133] The computer device includes a face detector; in this step, the computer device inputs the first image data into the face detector and outputs a face image of the sample user.
[0134] (2) Based on the face image, determine the facial feature points of the sample user.
[0135] The computer device includes a feature point recognition model; in this step, the computer device inputs the face image into the feature point recognition model and outputs the face feature points of the sample user.
[0136] (3) Based on the facial feature points, a set of action features of the sample user while driving a car is determined.
[0137] The motion feature includes the head swing amplitude, which can be represented by the rotation angles of the X-axis, Y-axis and Z-axis. Accordingly, step (3) can be implemented by the following steps (3-1) to (3-2), including:
[0138] (3-1) Determine the first coordinate information of the facial feature points in the world coordinate system and the second coordinate information of the facial feature points in the camera coordinate system.
[0139] (3-2) Based on the first coordinate information and the second coordinate information, the head swing amplitude of the sample user while driving the car is determined by the following formula 1.
[0140] Formula 1:
[0141] in, Represents the second coordinate information, represents the first coordinate information, [RT] represents the head swing amplitude. R represents the rotation matrix, T represents the translation matrix, and in one possible implementation, [RT] is used to represent the head swing amplitude of the sample user; in another possible implementation, R in [RT] is used to represent the head swing amplitude of the sample user.
[0142] The action features include the opening and closing distance of the left eye, the opening and closing distance of the right eye, the upper and lower opening and closing distance of the mouth, and the left and right opening and closing distance of the mouth; accordingly, step (3) can be implemented by the following steps (3-3) to (3-8), including:
[0143] (3-3) Determine multiple left eye key points, multiple right eye key points, multiple upper and lower mouth feature points, and multiple left and right mouth feature points from the facial feature points.
[0144] The computer device includes a key point detector; in this step, the computer device inputs the facial feature points into the key point detector, and outputs a plurality of left eye key points, a plurality of right eye key points, a plurality of upper and lower mouth feature points, and a plurality of left and right mouth feature points. The plurality of upper and lower mouth feature points include a plurality of upper mouth feature points and a plurality of lower mouth feature points, and the plurality of left and right mouth feature points include a plurality of left mouth feature points and a plurality of right mouth feature points.
[0145] (3-4) Determine the third coordinate information of multiple left eye key points in the camera coordinate system, the fourth coordinate information of multiple right eye key points in the camera coordinate system, the fifth coordinate information of multiple upper and lower mouth feature points in the camera coordinate system, and the sixth coordinate information of multiple left and right mouth feature points in the camera coordinate system.
[0146] In this step, the computer device can also determine the seventh coordinate information of multiple left eye key points in the world coordinate system, the eighth coordinate information of multiple right eye key points in the world coordinate system, the ninth coordinate information of multiple upper and lower mouth feature points in the world coordinate system, and the tenth coordinate information of multiple left and right mouth feature points in the world coordinate system.
[0147] (3-5) Based on the third coordinate information of multiple left eye key points, determine the opening and closing distance of the left eye of the sample user while driving a car.
[0148] The computer device determines multiple groups of left eye feature points based on multiple left eye key points, where a group of left eye feature points includes a first feature point and a second feature point, the first feature point is an upper feature point of the left eye, the second feature point is a lower feature point of the left eye, and the first feature point and the second feature point are two feature points at corresponding positions; for each group of feature points, the distance between the first feature point and the second feature point is determined based on the third coordinate information of the first feature point and the third coordinate information of the second feature point included in the group of feature points; based on the above method, the distances corresponding to the multiple groups of left eye feature points are determined, and based on the distances corresponding to the multiple groups of left eye feature points, the maximum distance is determined as the left eye opening and closing distance of the sample user when driving a car; or, based on the distances corresponding to the multiple groups of left eye feature points, the average distance is determined as the left eye opening and closing distance of the sample user when driving a car.
[0149] It should be noted that in step (3-4), the computer device obtains the seventh coordinate information of multiple left eye key points in the world coordinate system. In this step, the third coordinate information is replaced by the seventh coordinate information to determine the left eye opening and closing distance.
[0150] (3-6) Based on the fourth coordinate information of multiple right eye key points, determine the right eye opening and closing distance of the sample user while driving the car.
[0151] The computer device determines multiple groups of right eye feature points based on multiple right eye key points, where a group of right eye feature points includes a third feature point and a fourth feature point, where the third feature point is an upper feature point of the right eye, and the fourth feature point is a lower feature point of the right eye, and the third feature point and the fourth feature point are two feature points at corresponding positions; for each group of feature points, the distance between the third feature point and the fourth feature point is determined based on the fourth coordinate information of the third feature point and the fourth coordinate information of the fourth feature point included in the group of feature points; based on the above method, the distances corresponding to the multiple groups of right eye feature points are determined, and based on the distances corresponding to the multiple groups of right eye feature points, the maximum distance is determined as the right eye opening and closing distance of the sample user when driving a car; or, based on the distances corresponding to the multiple groups of right eye feature points, the average distance is determined as the right eye opening and closing distance of the sample user when driving a car.
[0152] It should be noted that in step (3-4), the computer device obtains the eighth coordinate information of multiple right eye key points in the world coordinate system. In this step, the fourth coordinate information is replaced by the eighth coordinate information to determine the right eye opening and closing distance.
[0153] (3-7) Based on the fifth coordinate information of multiple upper and lower mouth feature points, determine the upper and lower opening and closing distance of the sample user's mouth while driving the car.
[0154] The computer device determines multiple groups of upper and lower mouth feature points based on multiple upper and lower mouth feature points, where one group of upper and lower mouth feature points includes a fifth feature point and a sixth feature point, where the fifth feature point is a feature point on the upper side of the mouth, and the sixth feature point is a feature point on the lower side of the mouth, and the fifth feature point and the sixth feature point are two feature points at corresponding positions; for each group of feature points, the distance between the fifth feature point and the sixth feature point is determined based on the fifth coordinate information of the fifth feature point and the fifth coordinate information of the sixth feature point included in the group of feature points; based on the above method, the distances corresponding to the multiple groups of upper and lower mouth feature points are determined, and based on the distances corresponding to the multiple groups of upper and lower mouth feature points, the maximum distance is determined as the upper and lower opening and closing distance of the sample user's mouth when driving a car; or, based on the distances corresponding to the multiple groups of upper and lower mouth feature points, the average distance is determined as the upper and lower opening and closing distance of the sample user's mouth when driving a car.
[0155] One thing that needs to be explained is that in step (3-4), the computer device obtains the ninth coordinate information of multiple upper and lower mouth feature points in the world coordinate system. In this step, the fifth coordinate information is replaced by the ninth coordinate information to determine the upper and lower opening and closing distance of the mouth.
[0156] (3-8) Based on the sixth coordinate information of multiple left and right mouth feature points, determine the left and right opening and closing distance of the sample user's mouth while driving the car.
[0157] The computer device determines multiple groups of left and right mouth feature points based on multiple left and right mouth feature points, where a group of upper and lower mouth feature points includes a seventh feature point and an eighth feature point, where the seventh feature point is a feature point on the left side of the mouth, and the eighth feature point is a feature point on the right side of the mouth, and the seventh feature point and the eighth feature point are two feature points at corresponding positions; for each group of feature points, the distance between the seventh feature point and the eighth feature point is determined based on the sixth coordinate information of the seventh feature point and the sixth coordinate information of the sixth feature point included in the group of feature points; based on the above method, the distances corresponding to the multiple groups of left and right mouth feature points are determined, and based on the distances corresponding to the multiple groups of left and right mouth feature points, the maximum distance is determined as the left and right opening and closing distance of the sample user's mouth when driving a car; or, based on the distances corresponding to the multiple groups of upper and lower mouth feature points, the average distance is determined as the left and right opening and closing distance of the sample user's mouth when driving a car.
[0158] One thing that needs to be explained is that in step (3-4), the computer device obtains the tenth coordinate information of multiple left and right mouth feature points in the world coordinate system. In this step, the sixth coordinate information is replaced with the tenth coordinate information to determine the left and right opening and closing distance of the mouth.
[0159] Step 204: Based on the multiple groups of action features, determine the maximum value, minimum value, variance, mean and median of each action feature included in the multiple groups of action features, and obtain the action features of the sample user in the process of driving the car.
[0160] A group of action features includes head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, mouth upper and lower opening and closing distance, and mouth left and right opening and closing distance; in this step, the computer equipment determines the maximum value, minimum value, variance, mean and median of the head swing amplitude included in multiple groups of action features, determines the maximum value, minimum value, variance, mean and median of the left eye opening and closing distance included in multiple groups of action features, determines the maximum value, minimum value, variance, mean and median of the right eye opening and closing distance included in multiple groups of action features, determines the maximum value, minimum value, variance, mean and median of the upper and lower opening and closing distance of the mouth included in multiple groups of action features, determines the maximum value, minimum value, variance, mean and median of the left and right opening and closing distance of the mouth included in multiple groups of action features, and obtains the action features of the sample user in the process of driving a car.
[0161] Step 205: Based on the multiple frames of first voice data, multiple groups of sound features of the sample user while driving a car are determined, each group of sound features including pitch, sound intensity, formant, effective voice length, speaking speed and number of pauses.
[0162] Because it is necessary to find the correlation between the voice characteristics of the sample users when driving and the Big Five personality traits, it is necessary to develop an algorithm to solve the characteristic values of the sample users' voices when driving, such as pitch, intensity, resonance peak, effective speech length, speaking speed and number of pauses, and then solve the maximum value, minimum value, variance, mean and median of these characteristics to obtain the voice characteristics of the sample users.
[0163] The sound feature includes pitch; accordingly, step 205 can be implemented by the following steps (1) to (2), including:
[0164] (1) For each frame of first speech data, determine the fundamental frequency value of the first speech data.
[0165] The pitch is determined by the vibration frequency of the sound source; in this step, the computer device performs autocorrelation calculation on each frame of the first voice data, and estimates the fundamental frequency value of each frame of the first voice data through the result of the autocorrelation calculation.
[0166] (2) Based on the fundamental frequency value of the first voice data, the pitch of the sample user determined based on the first voice data is determined by the following formula 2:
[0167] Formula 2:
[0168] Wherein, h represents the pitch of the sample user determined based on the first voice data, and f represents the fundamental frequency value of the first voice data.
[0169] Sound features include sound intensity; sound intensity is the energy carried by sound propagation and is determined by the amplitude of the sound. This patent is calculated from the frequency domain Fourier transform and energy spectrum. Perform Fourier transform on each frame of data to convert the time domain signal into a frequency domain signal. Square the amplitude of the frequency domain signal to calculate the energy spectrum of each frame. The energy spectrum represents the energy of each frequency component. Calculate the sound intensity of each frame based on the energy spectrum. Accordingly, this step can be: for each frame of first voice data, perform Fourier transform on the first voice data to obtain the frequency domain signal of the first voice data, and the frequency domain signal includes multiple frequencies; for each frequency in the frequency domain signal, determine the square of the amplitude of the frequency to obtain the energy spectrum of the frequency; based on the energy spectrum of each frequency, determine the sound pressure value of the first voice data; based on the sound pressure value of the first voice data, determine the sound intensity of the sample user determined based on the first voice data by the following formula three:
[0170] Formula 3:
[0171] Among them, Lp represents the sound intensity of the sample user determined based on the first voice data, p represents the sound pressure value, and p0 represents the reference sound pressure value; the reference sound pressure value can be set and changed as needed. In the embodiment of the present application, the reference sound pressure value is not specifically limited; for example, the reference sound pressure value can be 20 micropascals.
[0172] Sound features include resonance peaks; resonance peaks are broad peaks or local maximum values in the spectrum. Resonance peaks are areas where energy is relatively concentrated in the sound spectrum. They reflect the physical characteristics of the vocal tract. The key to extracting resonance peaks is to estimate the maximum value in the spectrum envelope of natural speech. Resonance peak information is contained in the frequency envelope, so the methods for extracting resonance peak parameters include the cepstrum method, the LPC (linear predictive coding) method, and the HHT (Huang transform) method. In the embodiment of the present application, the cepstrum method is used as an example to determine the resonance peak; accordingly, step 205 can be implemented by the following steps (3) to (8), including:
[0173] (3) Pre-emphasis processing is performed on multiple frames of first voice data.
[0174] (4) Performing windowing and framing processing on the multiple frames of first voice data after pre-emphasis processing to obtain multiple frames of second voice data.
[0175] The frame length of each frame of second voice data in multiple frames of second voice data is not greater than the frame length of frame processing; and the frame length of frame processing can be set and changed as needed. In the embodiment of the present application, the frame length of frame processing is not specifically limited.
[0176] (5) Perform Fourier transform on the multiple frames of second voice data by the following formula 4 to obtain frequency domain signals of the multiple frames of second voice data:
[0177] Formula 4:
[0178] Among them, X i (n) represents the frequency domain signal of the second speech data of the i-th frame, x i (n) represents the second voice data of the i-th frame, N represents the frame length of the frame processing, and n represents the number of frames of the second voice data;
[0179] (6) Performing an inverse Fourier transform on the frequency domain signals of the multiple frames of second speech data to obtain an inverse Fourier transform sequence of the multiple frames of second speech data.
[0180] (7) Based on the dot sequence of the multi-frame second voice data, the envelope of the frequency domain signal of the multi-frame second voice data is determined by the following formula 5:
[0181] Formula 5:
[0182] Among them, H i (k) represents X i (n) envelope, h i (n) represents the product of the window function and the cepstrum sequence, N represents the frame length of the frame processing, and n represents the number of frames of the second speech data.
[0183] (8) For each second voice data, a maximum value is determined on the envelope of the frequency domain signal of the second voice data to obtain the resonance peak of the sample user determined based on the second voice data.
[0184] The sound features include the effective speech length; accordingly, step 205 may be: determine the difference between the duration of the multiple frames of first speech data and the silence duration to obtain the effective speech length. The sound features include speech rate, which is the vocabulary capacity contained in a unit of time; accordingly, step 205 may be: determine the ratio of the total duration of the multiple frames of first speech data to the number of words to obtain the speech rate. The sound features include the number of pauses; accordingly, step 205 may be: remove the time period of the beginning of silence and the time period of the end of silence from the multiple frames of first speech data, and determine the number of times the silence duration in the remaining speech segments is greater than the preset duration as the number of pauses. The preset duration can be set and changed as needed, and the preset duration is not specifically limited in the embodiment of the present application; for example, the preset duration can be 500 milliseconds.
[0185] Step 206: Based on the multiple groups of sound features, determine the maximum value, minimum value, variance, mean and median of each sound feature included in the multiple groups of sound features to obtain the sound features of the sample user when driving a car.
[0186] A group of sound features includes pitch, sound intensity, resonance peak, effective speech length, speaking speed and number of pauses; in this step, the computer device determines the maximum value, minimum value, variance, mean and median of the pitch included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the sound intensity included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the effective speech length included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the speaking speed included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the number of pauses included in multiple groups of sound features, and obtains the sound features of the sample user in the process of driving a car.
[0187] Step 207: Obtain the first driving data of the sample user driving the car within a recent preset time and the first vehicle data of the car, as well as the second driving data and the second vehicle data of the car within a recent preset mileage.
[0188] The preset duration can be two weeks, and the preset mileage can be 100 kilometers. Through the automobile data embedding point, the first driving data, the first vehicle data, the second driving data, and the second vehicle data of the sample user in the process of driving the car are collected, and transmitted to the big data platform by signal transmission and recorded and stored; in this step, the computer device obtains the first driving data, the first vehicle data, the second driving data, and the second vehicle data from the big data platform.
[0189] Step 208: Determine variable features corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data.
[0190] This step can be implemented by following the steps (1) to (5), including:
[0191] (1) Based on the first driving data, a first set of driving characteristics of the vehicle is determined, the first set of driving characteristics comprising charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average lateral speed, and maximum steering wheel speed.
[0192] (2) Based on the first vehicle data, a first set of vehicle characteristics of the vehicle is determined, where the first set of vehicle characteristics includes an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the vehicle.
[0193] (3) Based on the second driving data, determine a second set of driving characteristics of the vehicle, the second set of driving characteristics including charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average lateral speed, and maximum steering wheel speed.
[0194] (4) Based on the second vehicle data, determine a second set of vehicle characteristics of the vehicle, the second set of vehicle characteristics including an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the vehicle.
[0195] (5) Determining variable characteristics based on the first set of driving characteristics, the first set of vehicle characteristics, the second set of driving characteristics, and the second set of vehicle characteristics.
[0196] In one possible implementation, the first group of driving features, the first group of vehicle features, the second group of driving features, and the second group of vehicle features are combined into variable features. In another possible implementation, the first group of driving features and the second group of driving features are combined into driving features; when the driving features include two identical features, one of the two features is deleted; the first group of vehicle features and the second group of vehicle features are combined into vehicle features; when the vehicle features include two identical features, one of the two features is deleted; the driving features and the vehicle features are combined into variable features, thereby deleting identical features, thereby improving the accuracy of the determined variable features.
[0197] Step 209: Generate sample data based on the personality characteristics, action characteristics, sound characteristics and variable characteristics. The label of the sample data is the personality characteristics, and the characteristics of the sample data include action characteristics, sound characteristics and variable characteristics.
[0198] The number of sample users is multiple; therefore, the computer device generates multiple sample data based on the personality characteristics, action characteristics, sound characteristics and variable characteristics of the multiple sample users, and then executes step 210 to perform model training.
[0199] In a possible implementation, the computer device may also obtain vehicle information of a car driven by a sample user, the vehicle information including at least one of the model of the vehicle, the age of the vehicle, and the configuration information of the vehicle, and determine vehicle characteristics based on the vehicle information; in this step, the computer device generates sample data based on personality characteristics, action characteristics, sound characteristics, variable characteristics, and vehicle characteristics, the label of the sample data is personality characteristics, and the characteristics of the sample data include action characteristics, sound characteristics, variable characteristics, and vehicle characteristics.
[0200] In another possible implementation, the computer device may also obtain maintenance information of the car driven by the sample user, the maintenance information including maintenance time, maintenance content, etc.; determine maintenance features based on the maintenance information; in this step, the computer device generates sample data based on personality features, action features, sound features, variable features and maintenance features, the label of the sample data is personality features, and the features of the sample data include action features, sound features, variable features and maintenance features.
[0201] In another possible implementation, the computer device may also obtain usage information of the car driven by the sample user, the usage information including usage time (for example, commuting time or weekend time, etc.), usage duration (the duration required for commuting time or driving on weekends), etc.; determine usage characteristics based on the usage information; in this step, the computer device generates sample data based on personality characteristics, action characteristics, sound characteristics, variable characteristics and usage characteristics, the label of the sample data is personality characteristics, and the characteristics of the sample data include action characteristics, sound characteristics, variable characteristics and usage characteristics.
[0202] In another possible implementation, the computer device may also obtain accident information of the car driven by the sample user, the accident information including the time of the accident, the amount of compensation paid for the accident and the parts repaired for the accident; determine the accident characteristics based on the accident information; in this step, the computer device generates sample data based on personality characteristics, action characteristics, sound characteristics, variable characteristics and accident characteristics, the label of the sample data is personality characteristics, and the characteristics of the sample data include action characteristics, sound characteristics, variable characteristics and accident characteristics.
[0203] In another possible implementation, the computer device may also combine multiple selections of vehicle characteristics, maintenance characteristics, usage characteristics, and accident characteristics to generate sample data; for example, the computer device combines vehicle characteristics and maintenance characteristics to generate sample data; accordingly, the process may be: the computer device obtains vehicle information of a car driven by a sample user, and determines vehicle characteristics based on the vehicle information; obtains maintenance information of a car driven by a sample user, and determines maintenance characteristics based on the maintenance information; in this step, the computer device generates sample data based on personality characteristics, action characteristics, sound characteristics, variable characteristics, vehicle characteristics, and maintenance characteristics, the label of the sample data is personality characteristics, and the characteristics of the sample data include action characteristics, sound characteristics, variable characteristics, vehicle characteristics, and maintenance characteristics.
[0204] Step 210: Based on the sample data, a personality feature recognition model is trained by a machine learning algorithm, and the personality feature recognition model is used to recognize the personality features of the user.
[0205] The computer device constructs multiple machine learning models based on multiple machine learning algorithms, determines quality parameter values of the multiple machine learning models, selects a target machine learning model with the best quality from the multiple machine learning models based on the quality parameter values of the multiple machine learning models, and then trains the target machine learning model based on the sample data. The model obtained after the training is the personality trait recognition model.
[0206] Among them, the multiple machine learning algorithms can be multiple algorithms such as random forest, naive Bayes, support vector machine, decision tree and neural network. The step of the computer device determining the quality parameter values of multiple machine learning models can be: for each machine learning model, the computer device determines the accuracy, precision, recall rate, and comprehensive evaluation score of the machine learning model, and performs weighted summation of the accuracy, precision, recall rate, and comprehensive evaluation score of the machine learning model to obtain the quality parameter value of the machine learning model. In the embodiment of the present application, since the machine learning model with the best quality is selected for training, the accuracy of the personality feature recognition model obtained based on the machine learning model training can be improved.
[0207] In the embodiment of the present application, since the action characteristics, sound characteristics and variable characteristics of the user in the process of driving a car are associated with the personality characteristics to generate sample data, the personality characteristic recognition model trained based on the sample data can analyze the user's personality characteristics based on the user's driving behavior during the user's driving process, which not only expands the application scenarios of personality characteristic recognition, but also improves the accuracy of personality characteristic recognition.
[0208] Please refer to Figure 3 , which shows a flowchart of a method for identifying personality characteristics according to an exemplary embodiment of the present application. Figure 3 , the method comprising:
[0209] Step 301: The computer device obtains multiple frames of second image data and multiple frames of third voice data of the user to be identified in the process of driving a car.
[0210] In some embodiments, this step is similar to step 202 and will not be repeated here.
[0211] Step 302: The computer device determines multiple sets of motion features of the user to be identified in the process of driving a car based on multiple frames of second image data, each set of motion features including head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, upper and lower opening and closing distance of the mouth, and left and right opening and closing distance of the mouth.
[0212] In some embodiments, this step is similar to step 203 and will not be repeated here.
[0213] Step 303: The computer device determines the maximum value, minimum value, variance, mean and median of each action feature included in the multiple groups of action features based on the multiple groups of action features, and obtains the action features of the user to be identified in the process of driving a car.
[0214] In some embodiments, this step is similar to step 204 and will not be described again here.
[0215] Step 304: The computer device determines multiple groups of sound features of the user to be identified in the process of driving a car based on multiple frames of the third voice data, each group of sound features including pitch, sound intensity, resonance peak, effective voice length, speaking speed and number of pauses.
[0216] In some embodiments, this step is similar to step 205 and will not be repeated here.
[0217] Step 305: The computer device determines the maximum value, minimum value, variance, mean and median of each sound feature included in the multiple groups of sound features based on the multiple groups of sound features, and obtains the sound features of the user to be identified in the process of driving a car.
[0218] In some embodiments, this step is similar to step 206 and will not be described again here.
[0219] Step 306: The computer device obtains the third driving data of the to-be-identified user driving the car within a recent preset time period and the third vehicle data of the car, as well as the fourth driving data and the fourth vehicle data of the car within a recent preset mileage.
[0220] In some embodiments, this step is similar to step 207 and will not be repeated here.
[0221] Step 307: The computer device determines the variable features corresponding to the user to be identified based on the third driving data, the third vehicle data, the fourth driving data and the fourth vehicle data.
[0222] In some embodiments, this step is similar to step 208 and will not be described again here.
[0223] Step 308: The computer device inputs the action features, sound features and variable features of the user to be identified into the personality feature recognition model, and outputs the personality features of the user to be identified.
[0224] In one possible implementation, the computer device may also obtain vehicle information of a car driven by the user to be identified, the vehicle information including at least one of the model of the vehicle, the age of the vehicle, and the configuration information of the vehicle, and determine vehicle characteristics based on the vehicle information; in this step, the computer device inputs the action characteristics, sound characteristics, variable characteristics, and vehicle characteristics of the user to be identified into a personality characteristic recognition model, and outputs the personality characteristics of the user to be identified.
[0225] In another possible implementation, the computer device can also obtain maintenance information of the car driven by the user to be identified, the maintenance information including maintenance time, maintenance content, etc.; determine maintenance features based on the maintenance information; in this step, the computer device inputs the action features, sound features, variable features and maintenance features of the user to be identified into the personality feature recognition model, and outputs the personality features of the user to be identified.
[0226] In another possible implementation, the computer device may also obtain usage information of the car driven by the user to be identified, the usage information including usage time (for example, commuting time or weekend time, etc.), usage duration (the duration required for commuting or driving on weekends), etc.; determine usage characteristics based on the usage information; in this step, the computer device inputs the action characteristics, sound characteristics, variable characteristics and usage characteristics of the user to be identified into the personality characteristic recognition model, and outputs the personality characteristics of the user to be identified.
[0227] In another possible implementation, the computer device can also obtain accident information of the car driven by the user to be identified, the accident information including the time of the accident, the amount of compensation paid for the accident and the parts repaired for the accident; determine the accident characteristics based on the accident information; in this step, the computer device inputs the action characteristics, sound characteristics, variable characteristics and vehicle characteristics of the user to be identified into the accident characteristic recognition model, and outputs the personality characteristics of the user to be identified.
[0228] In another possible implementation, the computer device may also combine multiple selections of vehicle characteristics, maintenance characteristics, usage characteristics, and accident characteristics to identify personality characteristics; for example, the computer device combines vehicle characteristics and maintenance characteristics to generate sample data; accordingly, the process may be: the computer device obtains vehicle information of a car driven by a user to be identified, and determines vehicle characteristics based on the vehicle information; obtains maintenance information of a car driven by a user to be identified, and determines maintenance characteristics based on the maintenance information; in this step, the computer device inputs the action characteristics, sound characteristics, variable characteristics, vehicle characteristics, and maintenance characteristics of the user to be identified into a personality characteristic recognition model, and outputs the personality characteristics of the user to be identified.
[0229] In the embodiment of the present application, since the action characteristics, sound characteristics and variable characteristics of the user in the process of driving a car are associated with the personality characteristics to generate sample data, the personality characteristic recognition model trained based on the sample data can analyze the user's personality characteristics based on the user's driving behavior during the user's driving process, which not only expands the application scenarios of personality characteristic recognition, but also improves the accuracy of personality characteristic recognition.
[0230] Please refer to Figure 4 , which shows a block diagram of a personality feature recognition device shown in an exemplary embodiment of the present application. The device includes:
[0231] A first determination module 401 is used to determine the personality characteristics of the sample user based on the questionnaire survey information;
[0232] The first acquisition module 402 is used to acquire multiple frames of first image data and multiple frames of first voice data of a sample user in the process of driving a car;
[0233] The second determination module 403 is used to determine multiple groups of action features of the sample user in the process of driving the car based on the multiple frames of first image data, each group of action features including head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, mouth upper and lower opening and closing distance, and mouth left and right opening and closing distance;
[0234] The third determination module 404 is used to determine the maximum value, minimum value, variance, mean and median of each action feature included in the multiple action features based on the multiple action features, and obtain the action features of the sample user in the process of driving the car;
[0235] A fourth determination module 405 is used to determine multiple groups of sound features of the sample user in the process of driving the car based on the multiple frames of the first voice data, each group of sound features including pitch, sound intensity, formant, effective voice length, speech speed and number of pauses;
[0236] A fifth determination module 406 is used to determine the maximum value, minimum value, variance, mean and median of each sound feature included in the multiple groups of sound features based on the multiple groups of sound features, and obtain the sound features of the sample user in the process of driving the car;
[0237] The second acquisition module 407 is used to acquire the first driving data of the sample user driving the car within a recent preset time and the first vehicle data of the car, and the second driving data of the sample user driving the car within a recent preset mileage and the second vehicle data of the car;
[0238] A sixth determination module 408, configured to determine variable features corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data;
[0239] A generating module 409, for generating sample data based on personality characteristics, action characteristics, sound characteristics and variable characteristics, wherein the label of the sample data is the personality characteristics, and the characteristics of the sample data include action characteristics, sound characteristics and variable characteristics;
[0240] The training module 410 is used to train a personality feature recognition model based on sample data through a machine learning algorithm, and the personality feature recognition model is used to recognize the personality features of the user.
[0241] In a possible implementation, the sixth determination module 408 is used to determine a first group of driving characteristics of the car based on the first driving data, the first group of driving characteristics including the number of charging times, the number of driving times, the driving time, the maximum speed, the average speed, the maximum acceleration, the average acceleration, the average acceleration, the average acceleration, and the maximum steering wheel speed; determine a first group of vehicle characteristics of the car based on the first vehicle data, the first group of vehicle characteristics including the average remaining power, the maximum remaining power, and the minimum remaining power of the battery of the car; determine a second group of driving characteristics of the car based on the second driving data, the second group of driving characteristics including the number of charging times, the number of driving times, the driving time, the maximum speed, the average speed, the maximum acceleration, the average acceleration, the average acceleration, the average acceleration, the average acceleration, and the maximum steering wheel speed; determine a second group of vehicle characteristics of the car based on the second vehicle data, the second group of vehicle characteristics including the average remaining power, the maximum remaining power, and the minimum remaining power of the battery of the car; determine variable characteristics based on the first group of driving characteristics, the first group of vehicle characteristics, the second group of driving characteristics, and the second group of vehicle characteristics.
[0242] In another possible implementation, the sixth determination module 408 is used to combine the first group of driving characteristics and the second group of driving characteristics into driving characteristics; when the driving characteristics include two identical characteristics, delete one of the two characteristics; combine the first group of vehicle characteristics and the second group of vehicle characteristics into vehicle characteristics; when the vehicle characteristics include two identical characteristics, delete one of the two characteristics; and combine the driving characteristics and the vehicle characteristics into variable characteristics.
[0243] In another possible implementation, the second determination module 403 is used to perform face detection on each frame of first image data to obtain a face image of the sample user; based on the face image, determine the facial feature points of the sample user; based on the facial feature points, determine a set of action features of the sample user in the process of driving a car.
[0244] In another possible implementation, the action feature includes a head swing amplitude; the second determination module 403 is used to determine first coordinate information of a facial feature point in a world coordinate system and second coordinate information of the facial feature point in a camera coordinate system; based on the first coordinate information and the second coordinate information, the head swing amplitude of the sample user in the process of driving a car is determined by the following formula 1;
[0245] Formula 1:
[0246] in, Represents the second coordinate information, It represents the first coordinate information, and [RT] represents the amplitude of head swing.
[0247] In another possible implementation, the action features include a left eye opening and closing distance, a right eye opening and closing distance, a mouth upper and lower opening and closing distance, and a mouth left and right opening and closing distance;
[0248] The second determination module 403 is used to determine multiple left eye key points, multiple right eye key points, multiple upper and lower mouth feature points and multiple left and right mouth feature points from the facial feature points; determine the third coordinate information of the multiple left eye key points in the camera coordinate system, the fourth coordinate information of the multiple right eye key points in the camera coordinate system, the fifth coordinate information of the multiple upper and lower mouth feature points in the camera coordinate system and the sixth coordinate information of the multiple left and right mouth feature points in the camera coordinate system; based on the third coordinate information of the multiple left eye key points, determine the opening and closing distance of the left eye of the sample user in the process of driving the car; based on the fourth coordinate information of the multiple right eye key points, determine the opening and closing distance of the right eye of the sample user in the process of driving the car; based on the fifth coordinate information of the multiple upper and lower mouth feature points, determine the upper and lower opening and closing distance of the mouth of the sample user in the process of driving the car; based on the sixth coordinate information of the multiple left and right mouth feature points, determine the left and right opening and closing distance of the mouth of the sample user in the process of driving the car.
[0249] In another possible implementation, the sound feature includes pitch;
[0250] The fourth determination module 405 is used to determine the fundamental frequency value of the first voice data for each frame of the first voice data; based on the fundamental frequency value of the first voice data, determine the pitch of the sample user determined based on the first voice data by the following formula 2:
[0251] Formula 2:
[0252] Wherein, h represents the pitch of the sample user determined based on the first voice data, and f represents the fundamental frequency value of the first voice data.
[0253] In another possible implementation, the sound characteristics include sound intensity;
[0254] The fourth determination module 405 is used to perform Fourier transform on each frame of the first voice data to obtain a frequency domain signal of the first voice data, where the frequency domain signal includes multiple frequencies; for each frequency in the frequency domain signal, determine the square of the amplitude of the frequency to obtain an energy spectrum of the frequency; based on the energy spectrum of each frequency, determine the sound pressure value of the first voice data; based on the sound pressure value of the first voice data, determine the sound intensity of the sample user determined based on the first voice data by the following formula three:
[0255] Formula 3:
[0256] Wherein, Lp represents the voice intensity of the sample user determined based on the first voice data, p represents the sound pressure value, and p0 represents the reference sound pressure value.
[0257] In another possible implementation, the sound characteristics include formants;
[0258] The fourth determination module 405 is used to perform pre-emphasis processing on the multiple frames of first voice data; perform windowing and framing processing on the multiple frames of first voice data after pre-emphasis processing to obtain multiple frames of second voice data; and perform Fourier transform on the multiple frames of second voice data according to the following formula 4 to obtain frequency domain signals of the multiple frames of second voice data:
[0259] Formula 4:
[0260] Among them, X i (n) represents the frequency domain signal of the second speech data of the i-th frame, x i (n) represents the second voice data of the i-th frame, N represents the frame length of the frame processing, and n represents the number of frames of the second voice data;
[0261] Perform an inverse Fourier transform on the frequency domain signal of the multiple frames of second voice data to obtain an inverted Fourier transform sequence of the multiple frames of second voice data; based on the inverted Fourier transform sequence of the multiple frames of second voice data, determine the envelope of the frequency domain signal of the multiple frames of second voice data by the following formula 5:
[0262] Formula 5:
[0263] Among them, H i (k) represents X i (n) envelope, h i (n) represents the product of the window function and the cepstrum sequence, N represents the frame length of the frame processing, and n represents the number of frames of the second speech data;
[0264] For each second voice data, a maximum value is determined on the envelope of the frequency domain signal of the second voice data to obtain the formant of the sample user determined based on the second voice data.
[0265] In the embodiment of the present application, since the action characteristics, sound characteristics and variable characteristics of the user in the process of driving a car are associated with the personality characteristics to generate sample data, the personality characteristic recognition model trained based on the sample data can analyze the user's personality characteristics based on the user's driving behavior during the user's driving process, which not only expands the application scenarios of personality characteristic recognition, but also improves the accuracy of personality characteristic recognition.
[0266] It should be noted that the personality feature recognition device provided in the above embodiment only uses the division of the above functional modules as an example when performing personality feature recognition. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the personality feature recognition device provided in the above embodiment and the personality feature recognition method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.
[0267] Figure 5It is a structural block diagram of a computer device provided in an embodiment of the present application. Generally, the computer device 500 includes: a processor 501, a memory 502, a voice receiving device 503 and a controller 504. The processor 501 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 501 can be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 501 may also include a main processor and a coprocessor. The main processor is a processor for processing data in an awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 501 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.
[0268] The memory 502 may include one or more computer-readable storage media, which may be non-transitory. The memory 502 may also include a high-speed random access memory and a non-volatile memory, such as one or more disk storage devices and flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 502 is used to store at least one instruction, which is used to be executed by the processor 501 to implement the lighting control method provided in the method embodiment of the present application.
[0269] In some embodiments, the computer device 500 may further include: a peripheral device interface 505 and at least one peripheral device. The processor 501, the memory 502 and the peripheral device interface 505 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 505 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 506, an audio circuit 507 and a power supply 508.
[0270] The peripheral device interface 505 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 501 and the memory 502. In some embodiments, the processor 501, the memory 502, and the peripheral device interface 505 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 501, the memory 502, and the peripheral device interface 505 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.
[0271] The radio frequency circuit 506 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 506 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 506 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 506 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, and the like. The radio frequency circuit 506 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: a metropolitan area network, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 506 may also include circuits related to NFC (Near Field Communication), which is not limited in this application.
[0272] The audio circuit 507 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 501 for processing, or input them into the radio frequency circuit 506 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the computer device 500. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 501 or the radio frequency circuit 506 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 507 may also include a headphone jack.
[0273] The power supply 508 is used to power various components in the computer device 500. The power supply 508 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 508 includes a rechargeable battery, the rechargeable battery can support wired charging or wireless charging. The rechargeable battery can also be used to support fast charging technology.
[0274] Those skilled in the art will understand that Figure 5 The structure shown in the figure does not constitute a limitation on the computer device 500, and the computer device 500 may include more or less components than those shown in the figure, or combine some components, or adopt a different arrangement of components.
[0275] The embodiment of the present application also provides a computer-readable storage medium, in which at least one program code is stored, and the at least one program code is loaded and executed by a processor to implement the personality trait recognition method described in any of the above implementations. Optionally, the storage medium can be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium can be a ROM (Read-Only Memory), a RAM (Random Access Memory), a CD-ROM (Compact Disc Read-Only Memory), a magnetic tape, a floppy disk, and an optical data storage device.
[0276] The embodiments of the present application also provide a computer program product, which stores at least one program code, and the at least one program code is loaded and executed by a processor to implement the personality feature recognition method shown in the above embodiments.
[0277] In some embodiments, the computer program product involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located at one location, or on multiple computer devices distributed at multiple locations and interconnected by a communication network. Multiple computer devices distributed at multiple locations and interconnected by a communication network may constitute a blockchain system.
[0278] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.
[0279] The above description is only for the purpose of facilitating those skilled in the art to understand the technical solution of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A method for identifying personality traits, It is characterized in that The method comprises: Based on the questionnaire information, determine the personality characteristics of the sample users; Acquire multiple frames of first image data and multiple frames of first voice data of the sample user in the process of driving a car; Based on the multiple frames of first image data, determining multiple groups of action features of the sample user in the process of driving a car, each group of action features including head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, upper and lower opening and closing distance of the mouth, and left and right opening and closing distance of the mouth; Based on the multiple groups of action features, determine the maximum value, minimum value, variance, mean and median of each action feature included in the multiple groups of action features to obtain the action features of the sample user in the process of driving a car; Based on the multiple frames of first voice data, determining multiple groups of voice features of the sample user in the process of driving a car, each group of voice features including pitch, sound intensity, formant, effective voice length, speech speed and number of pauses; Based on the multiple groups of sound features, determine the maximum value, minimum value, variance, mean and median of each sound feature included in the multiple groups of sound features to obtain the sound features of the sample user in the process of driving the car; Acquire first driving data of the sample user driving a car within a recent preset time period and first vehicle data of the car, and second driving data of the sample user driving a car within a recent preset mileage and second vehicle data of the car; Determining variable features corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data; Generate sample data based on the personality feature, the action feature, the sound feature and the variable feature, wherein the label of the sample data is the personality feature, and the features of the sample data include the action feature, the sound feature and the variable feature; Based on the sample data, a personality feature recognition model is trained by a machine learning algorithm, and the personality feature recognition model is used to recognize the personality features of the user.
2. The method according to claim 1, It is characterized in that The determining, based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data, the variable features corresponding to the sample user includes: Determine a first set of driving characteristics of the vehicle based on the first driving data, the first set of driving characteristics comprising charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average lateral speed, and maximum steering wheel speed; Determine a first set of vehicle characteristics of the vehicle based on the first vehicle data, the first set of vehicle characteristics comprising an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the vehicle; Determining a second set of driving characteristics of the vehicle based on the second driving data, the second set of driving characteristics comprising charging times, driving times, driving time, maximum speed, average speed, maximum acceleration, average acceleration, average acceleration, average lateral speed, and maximum steering wheel speed; determining a second set of vehicle characteristics of the vehicle based on the second vehicle data, the second set of vehicle characteristics comprising an average remaining power, a maximum remaining power, and a minimum remaining power of a battery of the vehicle; The variable characteristics are determined based on the first set of driving characteristics, the first set of vehicle characteristics, the second set of driving characteristics, and the second set of vehicle characteristics.
3. The method according to claim 2, It is characterized in that The determining the variable characteristics based on the first set of driving characteristics, the first set of vehicle characteristics, the second set of driving characteristics, and the second set of vehicle characteristics comprises: combining the first set of driving characteristics and the second set of driving characteristics into a driving characteristic; When the driving characteristics include two identical characteristics, deleting one of the two characteristics; combining the first set of vehicle features and the second set of vehicle features into a vehicle feature; When the vehicle features include two identical features, deleting one of the two features; The driving characteristics and the vehicle characteristics are combined into the variable characteristics.
4. The method according to claim 1, It is characterized in that The determining, based on the multiple frames of first image data, multiple groups of action features of the sample user in the process of driving the car includes: For each frame of first image data, performing face detection on the first image data to obtain a face image of the sample user; Based on the facial image, determining facial feature points of the sample user; Based on the facial feature points, a set of action features of the sample user during driving a car is determined.
5. The method according to claim 1, It is characterized in that The sound characteristics include pitch; The determining, based on the multiple frames of first voice data, multiple groups of sound features of the sample user in the process of driving a car includes: For each frame of first speech data, determining a fundamental frequency value of the first speech data; Based on the fundamental frequency value of the first voice data, the pitch of the sample user determined based on the first voice data is determined by the following formula 2: Formula 2: Among them, h represents the pitch of the sample user determined based on the first voice data, and f represents the fundamental frequency value of the first voice data.
6. The method according to claim 1, It is characterized in that The sound characteristics include sound intensity; The determining, based on the multiple frames of first voice data, multiple groups of sound features of the sample user in the process of driving a car includes: For each frame of first voice data, performing Fourier transform on the first voice data to obtain a frequency domain signal of the first voice data, wherein the frequency domain signal includes multiple frequencies; For each frequency in the frequency domain signal, determine the square of the amplitude of the frequency to obtain an energy spectrum of the frequency; Determining a sound pressure value of the first speech data based on the energy spectrum of each frequency; Based on the sound pressure value of the first voice data, the voice intensity of the sample user determined based on the first voice data is determined by the following formula 3: Formula 3: Wherein, Lp represents the voice intensity of the sample user determined based on the first voice data, p represents the sound pressure value, and p 0 Indicates the reference sound pressure value.
7. A device for identifying personality traits, It is characterized in that The device comprises: A first determination module is used to determine the personality characteristics of sample users based on the questionnaire survey information; A first acquisition module, used to acquire multiple frames of first image data and multiple frames of first voice data of the sample user in the process of driving a car; A second determination module is used to determine multiple groups of action features of the sample user in the process of driving a car based on the multiple frames of first image data, each group of action features including head swing amplitude, left eye opening and closing distance, right eye opening and closing distance, upper and lower opening and closing distance of the mouth, and left and right opening and closing distance of the mouth; A third determination module is used to determine the maximum value, minimum value, variance, mean and median of each action feature included in the multiple groups of action features based on the multiple groups of action features, so as to obtain the action features of the sample user in the process of driving a car; a fourth determination module, configured to determine, based on the multiple frames of first voice data, multiple groups of voice features of the sample user in the process of driving a car, each group of voice features including pitch, sound intensity, formant, effective voice length, speech speed, and number of pauses; a fifth determination module, configured to determine, based on the multiple groups of sound features, a maximum value, a minimum value, a variance, a mean value, and a median value of each sound feature included in the multiple groups of sound features, and obtain the sound features of the sample user in the process of driving a car; A second acquisition module is used to acquire first driving data of the sample user driving a car within a recent preset time period and first vehicle data of the car, as well as second driving data of the sample user within a recent preset mileage and second vehicle data of the car; a sixth determination module, configured to determine a variable feature corresponding to the sample user based on the first driving data, the first vehicle data, the second driving data, and the second vehicle data; A generating module, configured to generate sample data based on the personality feature, the action feature, the sound feature and the variable feature, wherein the label of the sample data is the personality feature, and the features of the sample data include the action feature, the sound feature and the variable feature; The training module is used to train a personality feature recognition model based on the sample data through a machine learning algorithm, and the personality feature recognition model is used to recognize the personality features of the user.
8. A computer device, It is characterized in that The computer device includes a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the personality feature recognition method according to any one of claims 1 to 6.
9. A computer-readable storage medium, It is characterized in that The storage medium stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the personality feature recognition method according to any one of claims 1 to 6.
10. A computer program product, It is characterized in that The product stores at least one program code, and the at least one program code is used to be executed by a processor to implement the method for identifying personality characteristics as described in any one of claims 1 to 6.
Citation Information
Patent Citations
Driving mode adjusting method, system and device based on emotion recognition and medium
CN116691699A
Personality characteristic determination method and device
CN118269998A
Driving model training method, driver recognition method and apparatus, device, and medium
WO2019056471A1
